CN110440764A - A kind of secondary alignment methods of meter detection holder, device and equipment - Google Patents
A kind of secondary alignment methods of meter detection holder, device and equipment Download PDFInfo
- Publication number
- CN110440764A CN110440764A CN201910765582.5A CN201910765582A CN110440764A CN 110440764 A CN110440764 A CN 110440764A CN 201910765582 A CN201910765582 A CN 201910765582A CN 110440764 A CN110440764 A CN 110440764A
- Authority
- CN
- China
- Prior art keywords
- meter
- holder
- circle
- detection
- similarity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
- G01C11/02—Picture taking arrangements specially adapted for photogrammetry or photographic surveying, e.g. controlling overlapping of pictures
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
- G01C11/04—Interpretation of pictures
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/64—Computer-aided capture of images, e.g. transfer from script file into camera, check of taken image quality, advice or proposal for image composition or decision on when to take image
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/695—Control of camera direction for changing a field of view, e.g. pan, tilt or based on tracking of objects
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
This application discloses a kind of secondary alignment methods of meter detection holder, device and equipment, while carrying out the round reference meter with reference map of detection based on the detection of perceptual hash similarity, colouring information based on holder shooting figure carries out similarity detection, the accuracy that ensure that similarity detection solves the existing technical problem low using the existing recognition accuracy of Hough loop truss technology progress meter identification.
Description
Technical field
This application involves holder vision servo system technical fields more particularly to a kind of meter to detect the secondary alignment side of holder
Method, device and equipment.
Background technique
Intelligent Mobile Robot walks along inspection road and carries out equipment routing inspection, stops, adjusts after reaching predetermined monitoring position
With cradle head preset positions it will be seen that light detection instrument is directed at monitored equipment, it is seen that light detection instrument obtains ratio by optical zoom
Clear picture appropriate.Due to robot navigation's positioning accuracy, traveling attitude misalignment and cradle head preset positions positioning accuracy equal error
Combined influence, cause position of the imaging position of same observed object when each inspection in the entire image visual field have it is larger partially
Difference.
Current substation inspection road inspection detects round meter position using Hough loop truss technology, shoots
Scheming current meter central point and reference map, there are deviations with reference to the pixel of meter central point, but still can search background in searching for
There are drawback, the accuracy rate for causing meter to identify reduces more loop truss of circle and multiple similar meters.
Summary of the invention
The embodiment of the present application provides a kind of secondary alignment methods of meter detection holder, device and equipment, existing for solving
Some carries out meter using Hough loop truss technology and identifies the low technical problem of existing recognition accuracy.
In view of this, the application first aspect provides a kind of meter detection secondary alignment methods of holder, including following step
It is rapid:
101, the holder shooting figure under preset focal length is obtained;
102, search radius is determined based on meter radius to be measured, using holder shooting figure center searching as search rectangular frame
Suo Zhongxin, adjustment Hough the edge of the circle threshold value carry out Hough loop truss, and the detection for obtaining the first preset quantity required is round;
103, the similarity detection carried out based on perceptual hash algorithm round to the detection and the similarity based on color
Detection obtains satisfaction and the pixel of reference meter in reference map in the detection circle and is less than or equal to the first preset pixel
Similar circle;
If 104, the similar circular quantity be 1, execute step 105, if the similar circular quantity be 1 with
On, then to all described similar round according to the serial number sortord of the reference meter in reference map progress sequence
Number sequence retains the identical phase with the meter serial number to be measured according to serial number of the meter to be measured in the reference map
Like circle, step 105 is executed;
105, the similar circular center of circle is obtained, the center of circle is calculated and the pixel of the meter central point to be measured is inclined
Difference, if the pixel deviations are greater than or equal to critical value, according to the shooting angle of the corresponding adjustment holder of the pixel deviations.
Preferably, before step 103, further includes:
100, the parameter information of reference meter in reference map is obtained, the parameter information includes: reference map title, table
Count type, the meter number finally retained, reference meter in the positional relationship of the reference map finally retained, with reference to meter
Radius, the coordinate of x-axis and y-axis, perceptual hash similarity vector and the similarity vector based on color.
Preferably, before step 100 further include:
1001, the search rectangular frame of the preset size determined centered on reference map center adjusts Hough the edge of the circle threshold value
Hough loop truss is carried out, the reference for obtaining the second preset quantity required is round;
1002, the reference nearest apart from the reference map center is round as reference meter to be measured, it is based on
The similarity of perceptual hash algorithm detects and the similarity detection based on color, obtains and the reference meter pixel to be measured
Less than or equal to the reference meter of the second preset pixel;
If 1003, the quantity of the reference meter is 1,1004 are thened follow the steps, if the reference meter
Quantity be 1 or more, then obtain parameter information of the reference meter in the reference map, execute step 1004;
1004, the parameter information of the reference meter and the reference meter is stored in configuration file.
Preferably, the preset focal length is one times of focal length.
The application second aspect provides a kind of meter detection secondary alignment device of holder, comprises the following modules:
Module is obtained, for obtaining the holder shooting figure under preset focal length;
Search module is search with holder shooting figure center for determining search radius based on meter radius to be measured
The search center of rectangle frame, adjustment Hough the edge of the circle threshold value carry out Hough loop truss, obtain the detection of the first preset quantity required
It is round;
Similarity detection module, for carrying out the detection of the similarity based on perceptual hash algorithm and base to the detection circle
It is detected in the similarity of color, obtains satisfaction and the pixel of reference meter in reference map in the detection circle and be less than or wait
In the similar circle of the first preset pixel;
Judgment module, if being 1 for the similar circular quantity, triggering adjustment module, if the similar circular number
Amount is 1 or more, then to all described similar round according to the sequence of the serial number of the reference meter in the reference map
Mode carries out serial number sequence, is retained and the meter serial number phase to be measured according to serial number of the meter to be measured in the reference map
The same similar circle, triggers the adjustment module;
The adjustment module calculates in the center of circle and the meter to be measured for obtaining the similar circular center of circle
The pixel deviations of heart point, if the pixel deviations are greater than or equal to critical value, according to the corresponding adjustment holder of the pixel deviations
Shooting angle.
Preferably, further includes:
The reference module connecting with the similarity detection module, the reference module are joined for obtaining benchmark in reference map
The parameter information of meter is examined, the parameter information includes: reference map title, meter type, the meter number finally retained, benchmark
Coordinate, perceptual hash phase with reference to positional relationship of the meter in the reference map finally retained, the radius with reference to meter, x-axis and y-axis
Like degree similarity vector of the vector sum based on color.
Preferably, the reference module specifically includes:
Submodule is searched for, the search rectangular frame of the preset size for determining centered on reference map center adjusts Hough
The edge of the circle threshold value carries out Hough loop truss, and the reference for obtaining the second preset quantity required is round;
Similarity detection sub-module, for the reference nearest apart from the reference map center is round as base to be measured
Standard refer to meter, based on perceptual hash algorithm similarity detection and based on color similarity detection, obtain with it is described to be measured
Reference meter pixel is less than or equal to the reference meter of the second preset pixel;
Judging submodule triggers configuration submodule, if described if the quantity for the reference meter is 1
The quantity of reference meter is 1 or more, then obtains parameter information of the reference meter in the reference map, is touched
Send out configuration submodule described;
The configuration submodule, for by the parameter information of the reference meter and the reference meter
It is stored in configuration file.
The application third aspect provides a kind of meter detection secondary aligning equipment of holder, the equipment include processor with
And memory:
Said program code is transferred to the processor for storing program code by the memory;
The processor is used for any meter inspection according to the instruction execution first aspect in said program code
Survey the secondary alignment methods of holder.
The application fourth aspect provides a kind of computer readable storage medium, and the computer readable storage medium is used for
Program code is stored, said program code is for executing any meter detection secondary alignment side of holder described in first aspect
Method.
The aspect of the application the 5th provides a kind of computer program comprising computer instruction, which is characterized in that when its
When being run on computer, so that the computer executes any meter described in first aspect and detects the secondary alignment side of holder
Method.
As can be seen from the above technical solutions, the embodiment of the present application has the advantage that
In the application, provide a kind of meter detection secondary alignment methods of holder, comprising the following steps: 101, obtain it is preset
Holder shooting figure under focal length;102, search radius is determined based on meter radius to be measured, is search square with holder shooting figure center
The search center of shape frame, adjustment Hough the edge of the circle threshold value carry out Hough loop truss, obtain the detection circle of the first preset quantity required
Shape;103, the similarity detection carried out based on perceptual hash algorithm round to detection and the similarity based on color detect, and obtain
Meet the similar circle for being less than or equal to the first preset pixel to the pixel of reference meter in reference map in detection circle;
If 104, similar circular quantity is 1, step 105 is executed, if similar circular quantity is 1 or more, retains and joins with benchmark
The similar circle of meter same position sequence is examined, step 105 is executed;105, the similar circular center of circle is obtained, the center of circle and base are calculated
Standard refers to the pixel deviations of meter central point, if pixel deviations are greater than or equal to critical value, adjusts according to pixel deviations are corresponding
The shooting angle of holder.Method provided herein carries out Hough loop truss to the holder shooting figure got, obtains demand
Quantity it is round with meter to be measured detection of the same size, base is being carried out to the round reference meter with reference map of detection
While the detection of perceptual hash similarity, the colouring information based on holder shooting figure carries out similarity detection, ensure that similar
The accuracy for spending detection, after obtaining meeting the similar circle of similitude size, if similar circular quantity only has 1,
Similar round and pixel deviations of the reference meter in reference map are directly calculated, if pixel deviations are more than or equal to critical
Value is shot after then so that holder is precisely aligned meter to be measured according to the shooting angle of the corresponding adjustment holder of pixel deviations, if phase
Like circular quantity at 1 or more, then to all similar round serial number sequence sides according to the reference meter in reference map
Formula carries out serial number sequence, retains similar circle identical to meter serial number to be measured according to serial number of the meter to be measured in reference map,
Similar round and pixel deviations of the reference meter in reference map are calculated again, if pixel deviations are greater than or equal to critical value,
It is shot, is solved existing after then so that holder is precisely aligned meter to be measured according to the shooting angle of the corresponding adjustment holder of pixel deviations
Some carries out meter using Hough loop truss technology and identifies the low technical problem of existing recognition accuracy.
Detailed description of the invention
Fig. 1 is the process signal of one embodiment that a kind of meter provided by the present application detects the secondary alignment methods of holder
Figure;
Fig. 2 is the process signal for another embodiment that a kind of meter provided by the present application detects the secondary alignment methods of holder
Figure;
Fig. 3 is the structural representation of one embodiment that a kind of meter provided by the present application detects the secondary alignment device of holder
Figure;
Fig. 4 is the parameter information schematic diagram of reference meter in the reference map provided in the embodiment of the present application;
Fig. 5 is the more round meter reference map testing result figures provided in the embodiment of the present application;
Fig. 6 is the single round meter reference map testing result figure provided in the embodiment of the present application;
Fig. 7 is the more round meter testing result figures without similarity detection provided in the embodiment of the present application;
Fig. 8 is the more round meter testing results by similarity detection provided in the embodiment of the present application;
Fig. 9 is that the arrester Leakage Current table shooting figure without similarity screening provided in the embodiment of the present application detects
Result figure;
Figure 10 is that the arrester Leakage Current table shooting figure by similarity screening provided in the embodiment of the present application detects
Result figure.
Specific embodiment
In order to make those skilled in the art more fully understand application scheme, below in conjunction in the embodiment of the present application
Attached drawing, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described embodiment is only this
Apply for a part of the embodiment, instead of all the embodiments.Based on the embodiment in the application, those of ordinary skill in the art exist
Every other embodiment obtained under the premise of creative work is not made, shall fall in the protection scope of this application.
In order to make it easy to understand, referring to Fig. 1, one of a kind of meter detection secondary alignment methods of holder provided by the present application
Embodiment, comprising:
Holder shooting figure under step 101, the preset focal length of acquisition.
It should be noted that obtaining preset focal length by Intelligent Mobile Robot holder first in the embodiment of the present application
Under meter holder shooting figure to be measured, preset focal length can be selected as one times of focal length.
Step 102 determines search radius based on meter radius to be measured, using holder shooting figure center searching as search rectangular frame
Suo Zhongxin, adjustment Hough the edge of the circle threshold value carry out Hough loop truss, and the detection for obtaining the first preset quantity required is round.
It should be noted that the arbitrary point of the edge image of original image is corresponding to pass through this point in Hough loop truss technology
All possible circles indicated in the three-dimensional space center of circle and radius, the curve of a three-dimensional space is corresponded to, for multiple sides
Edge point, put it is more, these put corresponding three-dimensional space curve meet at any quantity it is more, then they pass through common circle on
Point it is more, similar can also judge whether a circle is detected with the method for same threshold value.The application is real
It applies in example, Hough circle search radius is defined with the radius of meter to be measured, the search rectangular frame centered on shooting figure center, search
The length and width of rectangle frame can select according to the actual situation, and the size than meter to be measured is big in principle, by gradually decreasing edge threshold
The mode of value adjusts detection threshold value, obtains the circle of quantity required, and the circle of quantity required should be greater than the number equal to meter to be measured
Amount.
Step 103 carries out the detection of the similarity based on perceptual hash algorithm and the similarity based on color to detection circle
Detection obtains the phase for meeting in detection circle and being less than or equal to the first preset pixel with the pixel of reference meter in reference map
Like circle.
It should be noted that in the embodiment of the present application, for all circles detected in holder shooting figure, need with
Target meter circle in reference map carries out the similarity comparison of pixel, the circle in holder shooting figure in similarity threshold
Meter as to be measured.The round using circular extraction is carried out by way of overflowing water filling of Hough loop truss result, then could
Carry out pixel similarity comparison.Unrestrained water completion method is a kind of specific color filling connected region, can be connected to picture by setting
The bound of element and mode of communicating reach the method for different filling effects, have selected the area being connected with seed point automatically
Domain takes up the ball the region and replaces with designated color.Unrestrained water filling obtains radius of circle and center location, table dependent on Hough loop truss
Round in meter and separate picture is that subsequent similarity-rough set is analyzed and handled.
In the embodiment of the present application, perceptual hash similarity detection is carried out first to Hough loop truss result, which extracts
Similarity vector based on grayscale image, the similar type vector by comparing two images obtain similarity value, specific steps
It is described as follows:
In order to remove holder shooting figure image details, only retain the essential informations such as structure/light and shade, abandon different rulers
Very little/ratio bring image difference reduces the picture size of holder shooting figure, by the size of image down to 8*8,64 in total
Image after diminution is switched to 64 grades of gray scales by pixel, i.e. all pixels point only has 64 kinds of colors in total, calculates all 64 pictures
The average gray of element, the gray scale of each pixel is compared with average gray, and the picture more than or equal to average value is denoted as
1, it is denoted as 0 less than average gray, the result that completeer 64 pixel grey scales obtain compared with average value is combined,
One 64 integers are just constituted, this 64 integer is exactly the fingerprint of this Zhang Yun's platform shooting figure.It should be pointed out that this 64
The combination order of position integer is not important, as long as guaranteeing that all images all use same order.Obtain holder shooting figure
Fingerprint after, compare two pictures in 64 place values, if different data be no more than 5, (the first preset pixel, digit can
Define according to the actual situation) illustrate that two images are much like, while may be arranged as different data be more than 10 (can be with
Define according to the actual situation), just illustrate that this is two different images.
Also need to carry out Hough loop truss result the similarity detection based on color in the embodiment of the present application.Every image
The histogram of distribution of color can be generated, if the histogram of two images very close to, so that it may think that they are much like.Appoint
A kind of what color is made of RGB (RGB) three primary colors, so share 4 histograms, i.e. three Zhang San's primary colors histograms
With the histogram being finally synthesizing.If every kind of primary colors can take 256 values, entire color space shares 16,000,000 kinds of face
Color (256 cube) compares histogram for this 1600 million colors, and calculation amount all too is big, it is therefore desirable to using letter
Change algorithm, be divided into four areas for 0~255: 0~63 is the 0th area, and 64~127 be the 1st area, and 128~191 be the 2nd area, 192~
255 be the 3rd area, this, which means that RGB respectively, 4 areas, may be constructed 64 kinds of combinations (i.e. 4 cube) in total.It is any
A kind of color necessarily belongs to one of this 64 kinds combinations, can thus count each and combine the pixel quantity for including, group
At 64 dimensional vectors, this vector is exactly that the characteristic value of this image or the similarity of referred to as " fingerprint " image reform into
Vector similarity can calculate the similarity of image according to m-cosine.
In the embodiment of the present application, it will meet and reference meter (circle of the meter i.e. to be measured in reference map in reference map
Shape) pixel be less than or equal to the first preset pixel similar circle.Reference map can be the meter holder to be measured shooting of standard
Figure, therefore, can by by it is similar it is round pixel in the picture and meter to be measured pixel progress phase in the picture
Compare like degree, retains pixel and be less than or equal to the similar round round as target of the first preset pixel.
If step 104, similar circular quantity are 1, step 105 is executed, if similar circular quantity is 1 or more,
Serial number sequence is carried out according to the serial number sortord of the reference meter in reference map to all similar circles, according to table to be measured
The serial number counted in reference map retains similar circle identical to meter serial number to be measured, executes step 105.Step 105 obtains phase
Like the circular center of circle, the pixel deviations in the center of circle and meter central point to be measured are calculated, if pixel deviations are greater than or equal to critical value,
According to the shooting angle of the corresponding adjustment holder of pixel deviations.
It should be noted that if the circular similar circular quantity of target only one, at this point it is possible to directly acquire the phase
Like the circular center of circle, the pixel deviations of center of circle image corresponding with reference meter central point are calculated, if pixel deviations are greater than
Or be equal to critical value, then it, should according to the corresponding adjustment holder of pixel deviations according to the shooting angle of the corresponding adjustment holder of pixel deviations
The method of shooting angle be the prior art, such as: image pixel dimensions are 576*720 (H*W), if orientation field angle is 1
Degree, 1 degree is equal to 720 pixels, and 0.001 degree is equal to 7.2 pixels;If orientation field angle is 0.5 degree, 0.5 degree is equal to 720
Pixel, 0.01 degree is equal to 14.4 pixels;If orientation field angle is 0.9 degree, 0.9 is equal to 720 pixels, and 0.01 degree is equal to
8 pixels, therefore without repeating in the embodiment of the present application.
If the circular similar circular quantity of target is 1 or more, then needing to retain and reference meter identical bits
It sets the similar circle of sequence, calculates the similar circular center of circle, then to calculate the center of circle corresponding with corresponding reference meter central point
Image pixel deviations, if pixel deviations are greater than or equal to critical value, according to the shooting of the corresponding adjustment holder of pixel deviations
Angle.Therefore, the accurate secondary alignment of Intelligent Mobile Robot holder is just realized.
If the circular similar circular quantity of target at 1 or more, needs to find out in all similar circles to be measured
The corresponding similar circle of meter, at this time, it may be necessary to all similar circles carry out carry out serial number sequence, serial number sortord with it is to be measured
The serial number sortord of reference meter of the meter in reference map is identical, by serial number and sequence of the meter to be measured in reference map
Number identical similar round similar circle as meter to be measured.Then calculate the similar circular center of circle, then calculate the center of circle with
The pixel deviations of the corresponding image of corresponding reference meter central point, if pixel deviations are greater than or equal to critical value, root
According to the shooting angle of the corresponding adjustment holder of pixel deviations.Therefore, the accurate secondary of Intelligent Mobile Robot holder is just realized
Alignment.
The meter detection secondary alignment methods of holder provided in the embodiment of the present application, carry out the holder shooting figure got
Hough loop truss obtains the round with meter to be measured detection of the same size of quantity required, in detection circle and reference map
Reference meter while carry out based on the detection of perceptual hash similarity, colouring information based on holder shooting figure carries out phase
It is detected like degree, ensure that the accuracy of similarity detection, after obtaining meeting the similar circle of similitude size, if circle of similitude
The quantity of shape only has 1, then similar round and pixel deviations of the reference meter in reference map is directly calculated, if pixel is inclined
Difference is greater than or equal to critical value, then so that holder is precisely aligned meter to be measured according to the shooting angle of the corresponding adjustment holder of pixel deviations
After shot, if similar circular quantity at 1 or more, to it is all it is similar circle according to the reference table in reference map
The serial number sortord of meter carries out serial number sequence, is retained and meter serial number phase to be measured according to serial number of the meter to be measured in reference map
Same similar circle, then similar round and pixel deviations of the reference meter in reference map are calculated, if pixel deviations are greater than
Or be equal to critical value, then it is carried out after so that holder is precisely aligned meter to be measured according to the shooting angle of the corresponding adjustment holder of pixel deviations
Shooting solves the existing technical problem low using the existing recognition accuracy of Hough loop truss technology progress meter identification.
In order to make it easy to understand, referring to Fig. 2, present invention also provides a kind of the another of meter detection secondary alignment methods of holder
One embodiment, comprising:
Holder shooting figure under step 201, the preset focal length of acquisition.
It should be noted that the step 201 in the embodiment of the present application is consistent with the step 101 in a upper embodiment, herein
No longer repeated.
Step 202 determines search radius based on meter radius to be measured, using holder shooting figure center searching as search rectangular frame
Suo Zhongxin, adjustment Hough the edge of the circle threshold value carry out Hough loop truss, and the detection for obtaining the first preset quantity required is round.
It should be noted that the step 202 in the embodiment of the present application is consistent with the step 102 in a upper embodiment, herein
No longer repeated.
Following steps 203 to step 206 can carry out simultaneously with step 101~102, and sequencing is not present.
The search rectangular frame of step 203, the preset size determined centered on reference map center adjusts Hough the edge of the circle threshold
Value carries out Hough loop truss, and the reference for obtaining the second preset quantity required is round.
Step 204 regard the reference circle nearest apart from reference map center as reference meter to be measured, is breathed out based on perception
The similarity detection of uncommon algorithm and the similarity based on color detect, and acquisition is less than or equal to reference meter pixel to be measured
The reference meter of second preset pixel.
If step 205, the quantity of reference meter are 1,206 are thened follow the steps, if the quantity of reference meter
It is 1 or more, then obtains parameter information of the reference meter in reference map.
The parameter information of reference meter and reference meter is stored in configuration file by step 206.
It should be noted that utilizing similarity analysis configuration baseline figure meter parameter information text to be measured in the embodiment of the present application
Part, in reference map, defining Hough loop truss search range (can be inclined according to picture caused in navigation accuracy allowed band
Shifting amount determines the search rectangular frame of preset size in conjunction with reference map center), in the circular outcome of Hough loop truss, range image
The nearest circle in center is to be measured with reference to meter, the circular perceptual hash and the similarity information based on color is extracted, by it
The circle that it is detected carries out similarity detection with reference to meter with to be measured, retains and detects circle up to standard, and by reference table to be measured
The Automatic generation of information of meter is simultaneously stored to configuration file, is done for circle detection in subsequent holder shooting figure and subsequent precisely align
Prepare.The information of configuration file includes the parameter information of reference meter in reference map, as shown in figure 4, parameter information includes:
Reference map title (first row in Fig. 4), meter type (meter type can be round meter or rectangular meter, the in Fig. 4
Two column), the meter number that finally retains (the third column in Fig. 4), reference meter is in the position of the reference map finally retained
Relationship (in Fig. 4 the 4th column), with reference to the radius (the 5th column in Fig. 4) of meter, the coordinate the (the in Fig. 4 the 6th of x-axis and y-axis
Column and the 7th column), perceptual hash similarity vector (in Fig. 4 the 8th column) and based on the similarity vector of color (after in Fig. 4
Continuous 64 column).
Step 207, the parameter information for obtaining reference meter in reference map.
Step 208 carries out the detection of the similarity based on perceptual hash algorithm and the similarity based on color to detection circle
Detection obtains the phase for meeting in detection circle and being less than or equal to the first preset pixel with the pixel of reference meter in reference map
Like circle.
If step 209, similar circular quantity are 1, step 210 is executed, if similar circular quantity is 1 or more,
Serial number sequence is carried out according to the serial number sortord of the reference meter in reference map to all similar circles, according to table to be measured
The serial number counted in reference map retains similar circle identical to meter serial number to be measured, executes step 210;
Step 210 obtains the similar circular center of circle, calculates the pixel deviations in the center of circle and reference meter central point, if
Pixel deviations are greater than or equal to critical value, then according to the shooting angle of the corresponding adjustment holder of pixel deviations.
It should be noted that as shown in figure 5, the reference map that Fig. 5 is more round meters (main transformer three-phase oil temperature gauge) detects knot
Fruit, testing result is three circles in search range, remaining testing result is a circle, because with reference meter radius
Similar circle has been filtered out by similarity comparison, such as the round paster of the label " B " in Fig. 6.
The case where being more round meters for configuration information:
The configuration information generated by reference map, can detect that multiple tables in search range.By taking Fig. 5 main transformer temperature oil meter as an example,
The first left circle table nearest apart from center picture is to be measured with reference to meter.In shooting figure, Hough loop truss is by gradually
Threshold value is reduced, detects multiple circles, as shown in fig. 7, the thinking then detected according to similarity, identification and reference map reference table
The circle that meter similarity is met the requirements detects three round oil temperature gauges, calls configuration information, that is, find the meter (left side to be measured
First round table), as shown in Figure 8.At this point, calculating the similar circular center of circle, then calculate the center of circle and corresponding reference meter
The pixel deviations of the corresponding image of central point adjust if pixel deviations are greater than or equal to critical value according to pixel deviations are corresponding
The shooting angle of holder.Therefore, the accurate secondary alignment of Intelligent Mobile Robot holder is just realized
The case where being single round meter for configuration information:
By taking arrester Leakage Current table as an example, calls when holder shooting figure Hough loop truss and searched with reference to meter radius
Rope obtains testing result as shown in Figure 9, then all results is carried out similarity comparison with circle of reference table one by one, will be non-circular
Meter filters out, and final detection result is as shown in Figure 10.At this point it is possible to directly acquire the similar circular center of circle, calculate the center of circle with
The pixel deviations of the corresponding image of reference meter central point, if pixel deviations are greater than or equal to critical value, according to pixel
The shooting angle of the corresponding adjustment holder of deviation.
A kind of meter provided herein detects the secondary alignment device of holder, comprises the following modules:
First obtains module 301, for obtaining the holder shooting figure under preset focal length.
Search module 302 is search square with holder shooting figure center for determining search radius based on meter radius to be measured
The search center of shape frame, adjustment Hough the edge of the circle threshold value carry out Hough loop truss, obtain the detection circle of the first preset quantity required
Shape.
Similarity detection module 303, for carrying out the detection of the similarity based on perceptual hash algorithm and base to detection circle
It is detected in the similarity of color, obtains to meet in detection circle and be less than or equal to the with the pixel of reference meter in reference map
The similar circle of one preset pixel.
Judgment module 304, if being 1 for similar circular quantity, triggering adjustment module 305, if similar circular quantity
It is 1 or more, then to all similar round according to the serial number sortord of the reference meter in reference map progress sequence
Number sequence, retains similar circle identical to meter serial number to be measured according to serial number of the meter to be measured in reference map, triggering adjusts
Module 305;
Module 305 is adjusted, for obtaining the similar circular center of circle, the center of circle is calculated and the pixel of meter central point to be measured is inclined
Difference, if pixel deviations are greater than or equal to critical value, according to the shooting angle of the corresponding adjustment holder of pixel deviations.
Further, further includes:
Second connect with similarity detection module obtains module 306, and the reference module is joined for obtaining benchmark in reference map
The parameter information of meter is examined, parameter information includes: reference map title, meter type, meter number, the reference finally retained
Positional relationship of the meter in the reference map finally retained, the radius with reference to meter, x-axis and y-axis coordinate, perceptual hash similarity
Similarity vector of the vector sum based on color.
Further, further includes:
Search module 307, the search rectangular frame of the preset size for being determined centered on reference map center adjust Hough
The edge of the circle threshold value carries out Hough loop truss, and the reference for obtaining the second preset quantity required is round.
Similarity detection module 308, for the reference nearest apart from reference map center is round as reference to be measured
Meter, the similarity detection based on perceptual hash algorithm and the similarity detection based on color, obtain and reference table to be measured
Count the reference circle that pixel is less than or equal to the second preset pixel.
Judgment module 309, if configuration submodule 3064 is triggered, if with reference to round for being 1 with reference to circular quantity
Quantity be 1 or more, then similar round carried out according to the serial number sortord of the reference meter in reference map to all
Serial number sequence retains similar circle identical to meter serial number to be measured according to serial number of the meter to be measured in reference map, and triggering is matched
Set submodule 3064;
Submodule 310 is configured, for the parameter information of reference meter and reference meter to be stored in configuration file.
Present invention also provides a kind of meters to detect the secondary aligning equipment of holder, and equipment includes processor and memory;
Program code is transferred to processor for storing program code by memory;
Processor is used to be implemented according to the secondary alignment methods of instruction execution meter detection holder above-mentioned in program code
Meter in example detects the secondary alignment methods of holder.
A kind of embodiment of computer readable storage medium is additionally provided in the application, computer readable storage medium is used for
Program code is stored, it is the meter in embodiment that said program code, which is used to execute the aforementioned meter detection secondary alignment methods of holder,
Detect the secondary alignment methods of holder.
A kind of computer program comprising computer instruction is additionally provided in the application, when run on a computer,
So that execute the meter detection holder in the meter above-mentioned detection secondary alignment methods embodiment of holder secondary right for the computer
Quasi- method.
In several embodiments provided herein, it should be understood that disclosed device and method can pass through it
Its mode is realized.For example, the apparatus embodiments described above are merely exemplary, for example, the division of the unit, only
Only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units or components can be tied
Another system is closed or is desirably integrated into, or some features can be ignored or not executed.Another point, it is shown or discussed
Mutual coupling, direct-coupling or communication connection can be through some interfaces, the INDIRECT COUPLING or logical of device or unit
Letter connection can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
It, can also be in addition, each functional unit in each embodiment of the application can integrate in one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of software functional units.
If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product
When, it can store in a computer readable storage medium.Based on this understanding, the technical solution of the application is substantially
The all or part of the part that contributes to existing technology or the technical solution can be in the form of software products in other words
It embodies, which is stored in a storage medium, including some instructions are used so that a computer
Equipment (can be personal computer, server or the network equipment etc.) executes the complete of each embodiment the method for the application
Portion or part steps.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory (full name in English: Read-Only
Memory, english abbreviation: ROM), random access memory (full name in English: Random Access Memory, english abbreviation:
RAM), the various media that can store program code such as magnetic or disk.
The above, above embodiments are only to illustrate the technical solution of the application, rather than its limitations;Although referring to before
Embodiment is stated the application is described in detail, those skilled in the art should understand that: it still can be to preceding
Technical solution documented by each embodiment is stated to modify or equivalent replacement of some of the technical features;And these
It modifies or replaces, the spirit and scope of each embodiment technical solution of the application that it does not separate the essence of the corresponding technical solution.
Claims (10)
1. a kind of meter detects the secondary alignment methods of holder, which comprises the following steps:
101, the holder shooting figure under preset focal length is obtained;
102, search radius is determined based on meter radius to be measured, using holder shooting figure center as in the search of search rectangular frame
The heart, adjustment Hough the edge of the circle threshold value carry out Hough loop truss, and the detection for obtaining the first preset quantity required is round;
103, the similarity detection carried out based on perceptual hash algorithm round to the detection and the similarity based on color detect,
It obtains in the detection circle and meets the phase for being less than or equal to the first preset pixel with the pixel of reference meter in reference map
Like circle;
If 104, the similar circular quantity is 1, step 105 is executed, if the similar circular quantity is 1 or more,
To all described similar round according to the serial number sortord of the reference meter in reference map progress serial number row
Sequence retains the circle of similitude identical with the meter serial number to be measured according to serial number of the meter to be measured in the reference map
Shape executes step 105;
105, the similar circular center of circle is obtained, the pixel deviations in the center of circle Yu the meter central point to be measured are calculated, if
The pixel deviations are greater than or equal to critical value, then according to the shooting angle of the corresponding adjustment holder of the pixel deviations.
2. meter according to claim 1 detects the secondary alignment methods of holder, which is characterized in that before step 103, also wrap
It includes:
100, the parameter information of reference meter in reference map is obtained, the parameter information includes: reference map title, meter class
Type, the meter number finally retained, positional relationship of the reference meter in the reference map finally retained, half with reference to meter
Diameter, the coordinate of x-axis and y-axis, perceptual hash similarity vector and the similarity vector based on color.
3. meter according to claim 2 detects the secondary alignment methods of holder, which is characterized in that also wrapped before step 100
It includes:
1001, the search rectangular frame of the preset size determined centered on reference map center, adjustment Hough the edge of the circle threshold value carry out
Hough loop truss, the reference for obtaining the second preset quantity required are round;
1002, the reference nearest apart from the reference map center is round as reference meter to be measured, based on perception
The similarity of hash algorithm detects and the similarity detection based on color, acquisition are less than with the reference meter pixel to be measured
Or the reference meter equal to the second preset pixel;
If 1003, the quantity of the reference meter is 1,1004 are thened follow the steps, if the number of the reference meter
Amount is 1 or more, then obtains parameter information of the reference meter in the reference map, executes step 1004;
1004, the parameter information of the reference meter and the reference meter is stored in configuration file.
4. meter according to claim 1 detects the secondary alignment methods of holder, which is characterized in that the preset focal length is one
Times focal length.
5. a kind of meter detection holder two surveys alignment device, which is characterized in that comprise the following modules:
Module is obtained, for obtaining the holder shooting figure under preset focal length;
Search module, for determining search radius based on meter radius to be measured, using holder shooting figure center as search rectangular
The search center of frame, adjustment Hough the edge of the circle threshold value carry out Hough loop truss, and the detection for obtaining the first preset quantity required is round;
Similarity detection module detects and for carrying out the similarity based on perceptual hash algorithm to the detection circle based on face
The similarity of color detects, and obtains to meet in the detection circle and is less than or equal to the with the pixel of reference meter in reference map
The similar circle of one preset pixel;
Judgment module, if being 1 for the similar circular quantity, triggering adjustment module, if the similar circular quantity is 1
More than a, then to all similar round serial number sortords according to the reference meter in the reference map into
The sequence of row serial number retains institute identical with the meter serial number to be measured according to serial number of the meter to be measured in the reference map
Similar circle is stated, the adjustment module is triggered;
The adjustment module calculates the center of circle and the meter central point to be measured for obtaining the similar circular center of circle
Pixel deviations, if the pixel deviations are greater than or equal to critical value, according to the bat of the corresponding adjustment holder of the pixel deviations
Take the photograph angle.
6. meter according to claim 5 detects the secondary alignment device of holder, which is characterized in that further include:
The reference module connecting with the similarity detection module, the reference module is for obtaining reference table in reference map
The parameter information of meter, the parameter information include: reference map title, meter type, meter number, the reference finally retained
Positional relationship of the meter in the reference map finally retained, the radius with reference to meter, x-axis and y-axis coordinate, perceptual hash similarity
Similarity vector of the vector sum based on color.
7. meter according to claim 6 detects the secondary alignment device of holder, which is characterized in that the reference module is specific
Include:
Submodule is searched for, the search rectangular frame of the preset size for determining centered on reference map center adjusts Hough round edge
Edge threshold value carries out Hough loop truss, and the reference for obtaining the second preset quantity required is round;
Similarity detection sub-module, for the reference nearest apart from the reference map center is round as benchmark to be measured ginseng
Meter is examined, the similarity detection based on perceptual hash algorithm and the similarity detection based on color obtain and the benchmark to be measured
It is less than or equal to the reference meter of the second preset pixel with reference to meter pixel;
Judging submodule triggers configuration submodule, if the benchmark if the quantity for the reference meter is 1
Quantity with reference to meter is 1 or more, then obtains parameter information of the reference meter in the reference map, triggers institute
State configuration submodule;
The configuration submodule, for the parameter information of the reference meter and the reference meter to be stored in
Configuration file.
8. a kind of meter detects the secondary aligning equipment of holder, which is characterized in that the equipment includes processor and memory:
Said program code is transferred to the processor for storing program code by the memory;
The processor is used to be detected according to the described in any item meters of instruction execution claim 1-4 in said program code
The secondary alignment methods of holder.
9. a kind of computer readable storage medium, which is characterized in that the computer readable storage medium is for storing program generation
Code, said program code require the described in any item meters of 1-4 to detect the secondary alignment methods of holder for perform claim.
10. a kind of computer program comprising computer instruction, which is characterized in that when run on a computer, so that institute
It states computer perform claim and requires the described in any item meter detection secondary alignment methods of holder of 1-4.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910765582.5A CN110440764B (en) | 2019-08-19 | 2019-08-19 | Meter detection cradle head secondary alignment method, device and equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910765582.5A CN110440764B (en) | 2019-08-19 | 2019-08-19 | Meter detection cradle head secondary alignment method, device and equipment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110440764A true CN110440764A (en) | 2019-11-12 |
CN110440764B CN110440764B (en) | 2021-07-09 |
Family
ID=68436473
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910765582.5A Active CN110440764B (en) | 2019-08-19 | 2019-08-19 | Meter detection cradle head secondary alignment method, device and equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110440764B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111310753A (en) * | 2019-12-27 | 2020-06-19 | 广东电网有限责任公司电力科学研究院 | Meter alignment method and device |
CN112560837A (en) * | 2020-12-02 | 2021-03-26 | 南方电网电力科技股份有限公司 | Reading method, device, equipment and storage medium of pointer instrument |
CN112561986A (en) * | 2020-12-02 | 2021-03-26 | 南方电网电力科技股份有限公司 | Secondary alignment method, device, equipment and storage medium for inspection robot holder |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120243741A1 (en) * | 2011-02-16 | 2012-09-27 | Siemens Corporation | Object Recognition For Security Screening and Long Range Video Surveillance |
CN107393270A (en) * | 2017-07-26 | 2017-11-24 | 河海大学常州校区 | A kind of portable vision inspection device and method for electric meter detection |
CN108256524A (en) * | 2018-01-24 | 2018-07-06 | 国家电网公司 | A kind of automatic reading method of multilist index formula instrument |
CN109034151A (en) * | 2018-06-27 | 2018-12-18 | 华中科技大学 | A kind of localization method for the identification of multiple pointer instruments |
CN109063634A (en) * | 2018-07-29 | 2018-12-21 | 国网上海市电力公司 | Using the method for hough-circle transform identification target object region electrical symbol in power monitoring |
-
2019
- 2019-08-19 CN CN201910765582.5A patent/CN110440764B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120243741A1 (en) * | 2011-02-16 | 2012-09-27 | Siemens Corporation | Object Recognition For Security Screening and Long Range Video Surveillance |
CN107393270A (en) * | 2017-07-26 | 2017-11-24 | 河海大学常州校区 | A kind of portable vision inspection device and method for electric meter detection |
CN108256524A (en) * | 2018-01-24 | 2018-07-06 | 国家电网公司 | A kind of automatic reading method of multilist index formula instrument |
CN109034151A (en) * | 2018-06-27 | 2018-12-18 | 华中科技大学 | A kind of localization method for the identification of multiple pointer instruments |
CN109063634A (en) * | 2018-07-29 | 2018-12-21 | 国网上海市电力公司 | Using the method for hough-circle transform identification target object region electrical symbol in power monitoring |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111310753A (en) * | 2019-12-27 | 2020-06-19 | 广东电网有限责任公司电力科学研究院 | Meter alignment method and device |
CN111310753B (en) * | 2019-12-27 | 2023-07-14 | 广东电网有限责任公司电力科学研究院 | Meter alignment method and device |
CN112560837A (en) * | 2020-12-02 | 2021-03-26 | 南方电网电力科技股份有限公司 | Reading method, device, equipment and storage medium of pointer instrument |
CN112561986A (en) * | 2020-12-02 | 2021-03-26 | 南方电网电力科技股份有限公司 | Secondary alignment method, device, equipment and storage medium for inspection robot holder |
Also Published As
Publication number | Publication date |
---|---|
CN110440764B (en) | 2021-07-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110440764A (en) | A kind of secondary alignment methods of meter detection holder, device and equipment | |
CN106934795B (en) | A kind of automatic testing method and prediction technique of glue into concrete beam cracks | |
CN109102547A (en) | Robot based on object identification deep learning model grabs position and orientation estimation method | |
CN105809146B (en) | A kind of image scene recognition methods and device | |
CN109345620A (en) | Merge the improvement ICP object under test point cloud method of quick point feature histogram | |
CN103994786B (en) | Image detecting method for arc ruler lines of pointer instrument scale | |
CN105612249B (en) | Image processing apparatus, program, information storage medium and image processing method | |
CN111210468A (en) | Image depth information acquisition method and device | |
CN109978949B (en) | Crop identification and feature point three-dimensional coordinate extraction method based on computer vision | |
CN109974721A (en) | A kind of vision winding detection method and device based on high-precision map | |
CN110245561A (en) | A kind of face identification method and device | |
CN109858547A (en) | A kind of object detection method and device based on BSSD | |
CN106340044A (en) | Camera external parameter automatic calibration method and calibration device | |
CN108960069A (en) | A method of the enhancing context for single phase object detector | |
CN106295512A (en) | Many correction line indoor vision data base construction method based on mark and indoor orientation method | |
CN105637343A (en) | Detection control device, program, detection system, storage medium and detection control method | |
CN107346414A (en) | Pedestrian's attribute recognition approach and device | |
CN106534659B (en) | A kind of camera focusing method and system | |
CN108292440A (en) | The method that the characteristic point of the calibrating pattern is identified in one group of candidate point in the image of calibrating pattern | |
CN109711241A (en) | Object detecting method, device and electronic equipment | |
CN109559344A (en) | Frame detection method, device and storage medium | |
CN110991297A (en) | Target positioning method and system based on scene monitoring | |
CN110363176A (en) | A kind of image analysis method and device | |
CN108805872A (en) | The detection method and device of product | |
CN110909617A (en) | Living body face detection method and device based on binocular vision |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |